unofficial mirror of guile-user@gnu.org 
 help / color / mirror / Atom feed
From: Roel Janssen <roel@gnu.org>
To: "Ludovic Courtès" <ludo@gnu.org>
Cc: guile-user@gnu.org
Subject: Re: Questions about the (web client) module.
Date: Fri, 21 Jul 2017 00:36:13 +0200	[thread overview]
Message-ID: <87h8y691nm.fsf@gnu.org> (raw)
In-Reply-To: <878tjjru96.fsf@gnu.org>

Hi Ludo’,

Thanks for your response.

Ludovic Courtès writes:

> Hi Roel,
>
> Roel Janssen <roel@gnu.org> skribis:
>
>> When I use http-post, and I want to change the HTTP header called
>> "Content-Type", I seem to need to spell it as "content-type" in the
>> #:headers part of the 'http-post' section.  Other headers like "Accept"
>> do not seem to follow the same lowercase style route.
>>
>> More confusingly, using something like:
>>   #:headers '((Content-Type . "text/csv"))
>>
>> leads to outputting the "Content-Type" header twice.
>> Why is "content-type" special?
>
> ‘sanitize-request’ in (web client) adds a ‘Content-Type’ header if there
> is none.

But 'Content-Type' and 'content-type' is not the same.. Adding a
'Content-Type' header results in the request having two 'Content-Type'
headers.  So it doesn't add 'Content-Type' when it isn't there, it
converts 'content-type' to 'Content-Type', so it adds 'Content-Type'
when 'content-type' isn't there.  This seems wrong to me.

>
> As for lower-case, perhaps the ‘request’ procedure in (web client)
> should automatically convert to lower-case, or perhaps we should simply
> clarify the documentation here.

Or perhaps do not treat some HTTP headers different than others, and
just keep it 'Content-Type' (as used in HTTP) instead of 'content-type'.

>
>> Then my next question is about "multipart/form-data" content types.
>> My code looks like this:
>>
>>   #:headers `((content-type . ,(string-append
>>                                 "multipart/form-data; boundary=" boundary))
>>               (Accept . "*/*"))
>>
>> But that does not work:
>>   scheme@(guile-user)> 
>>   web/request.scm:184:10: In procedure build-request:
>>   web/request.scm:184:10: Bad request: Bad value for header content-type: "multipart/form-data; boundary=..."
>>
>> This is, however, a valid Content-Type.
>
> What’s the value of ‘boundary’?  At first sight it looks good to me:

The value I use is "guile-virtuoso-driver".  All of the below
functions work as expected with this value.

Could it be that something else is wrong?
Here's the entire code I use:

--8<---------------cut here---------------start------------->8---
(define* (sparql-query-post query
                       #:key
                       (host "localhost")
                       (port 8890)
                       (graph "")
                       (type "json"))
  (let ((base-uri (string-append "http://" host ":" (number->string port) ))
        (boundary "guile-virtuoso-driver"))
    (let ((body (string-append
                 (format #f
                  "--~a~%Content-Disposition: form-data; name=~s~%~%text/csv~%"
                  boundary "format")
                  (format #f
                  "--~a~%Content-Disposition: form-data; name=~s~%~%~a~%--~a~%"
                  boundary "query" query boundary))))
      (http-post
       (string-append base-uri "/sparql")
       #:body body
       #:headers `((content-length . ,(string-length body))
                   (content-type . "multipart/form-data; boundary=guile-virtuoso-driver")
                   (Accept . "*/*"))))))
  (define %query-get-all-triplets
  "SELECT * FROM <http://localhost:8890/packages>
WHERE
{
  ?s ?p ?o
}")

(sparql-query-post %query-get-all-triplets)
--8<---------------cut here---------------end--------------->8---

Which results in:

--8<---------------cut here---------------start------------->8---
web/request.scm:184:10: In procedure build-request:
web/request.scm:184:10: Bad request: Bad value for header content-type: "multipart/form-data; boundary=guile-virtuoso-driver"
--8<---------------cut here---------------end--------------->8---

Thanks!

Kind regards,
Roel Janssen




  reply	other threads:[~2017-07-20 22:36 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-20 10:53 Questions about the (web client) module Roel Janssen
2017-07-20 15:41 ` Ludovic Courtès
2017-07-20 22:36   ` Roel Janssen [this message]
2017-07-21  4:49     ` Mark H Weaver
2017-07-21  5:05     ` Mark H Weaver

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: https://www.gnu.org/software/guile/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87h8y691nm.fsf@gnu.org \
    --to=roel@gnu.org \
    --cc=guile-user@gnu.org \
    --cc=ludo@gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).